The growth in the amount of waste from industry and the urban environment has motivated research to be carried out worldwide with a view to its use in infrastructure works. Rio Grande do Norte is a Brazilian state where the mineral scheelite is being explored, a source for obtaining the chemical element Tungsten, a rare metal in the earth's crust and which has important industrial applications. The processing of scheelite causes a high generation of waste, as only 0.8% of the material extracted from the mines is used. The main environmental damages involved in scheelite extraction are soil erosion, atmospheric, noise and visual pollution resulting from the formation of waste piles. Thus, the objective of this work is to develop a mathematical model of mixtures composed of soil samples from deposits in three regions of the state of Rio Grande do Norte, characterized by Barreto (2020), and scheelite residues from this state, characterized by Souza (2019). The technical evaluations of scheelite and soil compositions were carried out through numerical analysis, which worked with increments of 20%, 40% and 60% of scheelite residues, in mass, associated with the three soils. The results showed that the mixtures with the best granulometric compositions for pavement base layers were composed by Soil 01 with 20% or 40% of residue (AASHTO and ASTM), or by Soil 01 with 20% of residue (DNIT).